Large Intelligent Surfaces for Energy Efficiency in Wireless Communication

Chongwen Huang, Alessio Zappone, George C. Alexandropoulos, Mérouane Debbah, Chau Yuen

Introduction

As the telecommunications industry transitions toward 6G networks, the demand for massive data rates has led to a significant increase in energy consumption. While Massive MIMO technologies offer improvements, they face diminishing returns due to hardware costs and power leakage. Traditionally, Amplify-and-Forward (AF) relays have been used to extend coverage to "dead zones." However, these active devices require power-hungry amplifiers that also amplify background noise.

This research proposes a paradigm shift from active relaying to Reconfigurable Intelligent Surfaces (RIS). An RIS is a passive meta-surface that "shapes" the wireless environment by reflecting incoming waves with specific phase shifts, rather than amplifying them.

System model of a multi-user wireless network assisted by a Reconfigurable Intelligent Surface (RIS).

Problem Statement and Methodology

The primary challenge lies in coordinating the base station's transmit power and the RIS's reflection phases to maximize Energy Efficiency (EE)—the ratio of the sum data rate to total power consumed—while maintaining a minimum quality of service for all users.

The authors address this non-convex optimization problem through several key innovations:

  1. Realistic Power Modeling: Unlike previous models, this work accounts for the static power dissipated by each reflecting element, which varies based on phase-shifting resolution.
  2. Alternating Maximization: To handle the mathematical complexity, the authors employ an iterative approach, alternating between optimizing the base station power and the RIS phase shifts.
  3. Algorithmic Approaches: Two primary methods are evaluated: Gradient Descent (GD) with a quadratic approximation for step size, and Sequential Fractional Programming (SFP) using majorization-minimization to handle unit-modulus constraints.

Simulation and Evaluation

The framework was evaluated using Monte Carlo simulations based on 3GPP propagation models for outdoor urban environments. The setup considers a scenario where users are located behind obstacles, necessitating the use of the RIS for reliable communication.

Simulation setup depicting the placement of the Base Station, RIS, and users behind an obstacle.

Results were averaged over 1,000 independent trials involving varying user positions and channel conditions within a 100m x 100m area. The hardware specifications included real-world power consumption values for phase shifters, ranging from 1.5mW to 7.8mW.

Key Insights and Results

  • 300% Efficiency Gain: In realistic outdoor scenarios, the RIS-based system achieved up to three times the energy efficiency of traditional AF relay systems.
  • Passive Advantage: By avoiding power amplifiers, RIS eliminates noise amplification, a significant drawback of active relaying.
  • The "Optimal N" Trade-off: There is a critical threshold for the number of reflecting elements ($N$). While more elements increase the data rate, the cumulative static power consumption eventually leads to a decline in overall energy efficiency.
  • Saturation Effects: EE gains tend to saturate at high transmit powers as the base station's power consumption becomes the dominant factor in the total energy budget.

Conclusion and Future Directions

The study demonstrates that RIS is a highly promising technology for green wireless communications. However, several practical hurdles remain. The current model assumes perfect Channel State Information (CSI), which is difficult to obtain for passive surfaces. Future research should focus on robust channel estimation, the impact of finite phase resolution, and the challenges posed by high-speed user mobility.

Summary updated

BibTeX
@article{huang2019large,
  author = {Chongwen Huang and Alessio Zappone and George C. Alexandropoulos and Mérouane Debbah and Chau Yuen},
  title = {{Large Intelligent Surfaces for Energy Efficiency in Wireless Communication}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2019},
  doi = {10.1109/twc.2019.2922609},
}
Figure